TV day at Dulwich - 10th Sept

[ppppenguin]

I've been in correspondence with Darryl about various Aurora related matters. I mentioned the hum problem we'd been having on BBC1 and he related this story. He uses term "line capacitor", we would say aerial isolator capacitor in the UK. However you look at it the results were not pretty. I've attached the photos he sent me.

So we recently had two of the original Multi-Standard converters fail at the ETF. What happened was very similar to your roll bar issue at the Museum, but with much more devastating consequences. Steve was having an issue in the British room and noticed the converter no longer appeared to be working. Without thought, he grabbed one from another room, plugged it in, and BOOM! It turned out to be a failed line capacitor in one of the pre-war sets (not sure which one) that put mains on the antenna shield. He did not have a good ground on this shield, expect where it ran through the converter to upstream equipment. This was the result:

Most of you won't be famiiair with the original Aurora shown here. It could produce many different output standards at the touch of a switch but it didn't have a modulator. About a year later Darryl introduced the well know SCRF converter that many of us use.

Test of my review of the original Aurora from 2005:
http://www.borinsky.plus.com/aurora_conv...review.htm


My review of the SCRF405A in 2006. This has gone through 4 major hardware revisions. The basic video performance is to the same excellent quality on all of them. Over 700 have been sold, mostly to the UK.
http://www.borinsky.plus.com/aurora2_con...lished.pdf

[Refugee]

I would not be betting on those converters ever working again. The plasma has blasted all the way across to the processor chip.

[ppppenguin]

Darryl said they were repaired!

The shield lugs and three layers of ground foil on the pcb’s vaporized. The black coating on the boards was imbedded into plastic and steel parts. (notice the shields on the mini-DIN connectors)


Now the really amazing part; beside having to change out the destroyed connectors and repair the lost foils connections, the only electrical damage to the boards were the 75 ohm terminators on the input and output lines! There was no damage to the output opamps or the input decoders. I am at a loss for how this is possible.

[Murphyv310]

Hi Jeffrey.
Well done to Darryl for repairing them, he is an excellent engineer and it good to see the original converters still being used and fixed!!
I still use mine on occasions and have threatened many a time to do some experiments on the lower line standards.
My old Decca 1000 projection set will resolve the Baird 240 line system with flicker and a bit of right hand side cramping, It's actually quite eeary to watch in a strange kind of way, certainly not of any entertainment value with the flicker.
Give Darryl a pat on the back from me

[Gryphon]

I may, perhaps, be able to claim some modest expertise with the B18T/LV20, having now restored quite a good number of them.

As the circuit shows, the B18T/LV20 was designed to be used with a balanced feeder.  I believe that there was an official Pye mod which was issued to convert it for use with a coax feeder, but I've never seen it.  I assume that this will have involved replacing T1A.

What I have seen are all sorts of bodges from the past. Suffice it to say that some of these were more hygienic than others.

After quite a bit of thought and experimentation, the conclusion I eventually arrived at was that the only really satisfactory way of using one of these sets with a coax feeder is, having checked that the mod. done is not the official Pye one,  to undo whatever has been done in the past, put the circuit back to "as designed" and use an external balun.

This is easily made by making two windings each of 10 turns of a convenient gauge of ECW (I find about 28 or 30 swg is a convenient compromise between ease of handling and fragility) on a miniature VHF ferrite toroidal core.  If the windings are spaced diametrically opposite each other, this overcomes the problem that mains is more than most wire enamels are designed to withstand.  The resulting coupling isn't, perhaps, as tight as is ideal, but this isn't a problem in a modern setting where, usually, more than enough signal is available from the r.f. source.  Superglue is very useful to stop the windings "coming undone".  It is important to use a grade of ferrite which is specified for VHF applications, however. HF ferrites are usually more lossy at 45MHz than is ideal.

I used to make mine in the plastic boxes which 35mm stills film cassettes were supplied in, though anything similar will do.  

I believe that there is a problem of a similar nature with the BV20 (receiver strip tuned to Sutton Coldfield).

[ppppenguin]

Thanks for giving details of how to do the job properly, I cannot fault your reasoning or methods. However I feel it's not really necessary if the set is still unmodified. Merely connecting a 75R co-ax feed to the balanced input will give ample signal on any system we're now likely to use. The mismatch will also be unimportant. To give you an idea, the Vintage WIreless Museum's distribution system is designed to give +6dBmV at each outlet which is equal to 2mV. With a tolerance range of 1mV to 3mV. Most of us will be using an Aurora as a source of 405 signals. Even when split 8 ways the signal will still be of the same order. When the signal is that strong I don't think we need to worry about losing some through bad matching.

Any reflections due to bad matching are unlikely to give trouble. They will tend to absorbed at the sending end so the "3rd time aound" reflection will be negligible. If there genuinely is a reflection problem and some more signal can be sacrificed the old trick of an attenuator in the feed is a good one. A 6dB pad will reduce the wanted signal by 6dB while the 3rd time around reflection will have passed through the attenuator 3 times and hence be 18dB down, or 12dB lower than the wanted signal.

[ppppenguin]

I ought to add a note about safety when using Auroras, modulators etc with old TVs. Many sets have their chassis connected to one side of the mains. The aerial isolator in the set keeps mains away from the aerial. Unfortunately the isolators can fail with unpleasant results as we've seen above. Even worse, if nothing is earthed and the entire aerial distribution system goes live you've got a nasty shock hazard.

Hence I recommend:

1: Solidly earth the co-ax outer of an aerial distribution system. Mains earth is fine for this. If an errant set puts live mains on the co-ax it will at least blow a fuse.

2: Use an RCD in the power feed to the sets. Always sensible. Will trip on lesser faults that wouldn't blow a fuse.

3: Ensure that the chassis of each set is connected to neutral. Mark connectors that are reversible. At least this will reduce the likelihood of damage if something goes wrong.

For larger systems and workbenches consider using additional isolators. There's a good reason why every Virgin cable installation has one of these where their cable enters the house. At the Museum we use one in the feed from the sheds to the house to ensure that a fault at one end doesn't affect the other. We also use one in the feed to the workbench so that a duff set or accident there doesn't affect anything else. A DIY isolator, along the lines of Gryphon's balun will be fine if nothing else is available. Or even a couple of Y rated 1000p caps in a plastic 35mm film can.

[Mike Watterson]

All good advice.

It's an issue even with modern systems!
Sharing TV, Satellite, Intercom, HiFi, even networks between neighbours should be via a proper earthed barrier. There are documents on safety and barrier earthing and isolation for "Multi-Dwelling" installations (usually apartments but not always). Sharing a satellite dish may exceed the isolation rating of poorly specified  SMPSU input filter capacitors due to neighbouring houses on different phases. The satellite receivers and TVs often have no earth.

The LNBs though should always be isolated from the dish, which sometimes ought to have a separate local earth spike, but often isn't explicitly earthed.

[Cardigan]

Gryphon's post had me thinking: I wonder if it would be possible to adapt the 300 -> 75 Ohm baluns you can buy for VHF/FM tuners? One would assume that halving the number of turns on the 300 Ohm side would do the trick.

[Mike Watterson]

I wonder if it would be possible to adapt the 300 -> 75 Ohm baluns you can buy for VHF/FM tuners?

Not worth the effort, many are poor quality and may give poor isolation. A "braid breaker" sold to reduce CB radio interference likely is better but I'd not trust it either. The ferrite beads suitable for 40MHz up are cheap. I'd use PTFE insulated wire rather than enamel to get the isolation, or 1kV XY rated ceramic capacitors (which is what is often inside set-top aerials and late model live chassis, or both. PVC tube or old broad felt tip marker case.